Image Denoise Using Auto-Adapted Empirical Mode Decomposition Algorithm
Lingfei Liang, Ziliang Ping · 2009
In this paper, a new image denoise method is presented. The main contribution of our approach is to apply an auto-adapted empirical mode decomposition algorithm (AAEMD) to remove the noise. By means of AAEMD, the image can be decomposed into a number of intrinsic mode function (IMF), and the histogram of IMF observes the normal distribution, whether the original image has noise or not. Here, histogram matching is applied for denoise in the first few IMF images. In other words, the histogram of the new IMF images after Appling histogram matching observes the special normal distribution. At last, the modified and unmodified IMF images are added up to get the denoise image. Experiments prove that the novel algorithm is efficient in image denosise and better than current algorithms.